I learned tons building a household management app with Claude

A few months ago, I decided to take my household management to the next level. I’d already gotten pretty good at keeping up with the basic chores, but I needed a system to track our family calendar, family goals, and household inventory. Plus, I wanted a way to customize the experience for my family, pets, and the actual number of rooms in my house.

I tried several existing household apps, and a few came close to what I was looking for. Still, I wanted something completely tailored to my situation, not just something “close enough.” That got me thinking: could I make my own solution? Plenty of people are already using AI tools to learn programming, or for simple vibe coding, so I dove in with Claude to see what I could build, what walls I’d hit, and whether this could be a better fit for my needs.

There were a few ways to go about this, but I started with the simplest: a progressive web app. Below, I’ll walk through how I built it, what it does that other solutions can’t, and where this approach falls short.

You might wonder why I didn’t just use one of the many chore or household management apps that I tried out. I came close, actually. I played around with Cozi Family Organizer, OurHome, Sweepy, and a few other apps, trying to find the perfect solution. Some of these apps were nearly what I was looking for, but there was always something missing.

Sweepy has a highly customizable chores list that lets you add unique tasks and rooms to truly tailor it to your home, plus a decent calendar system. What you won’t find is a household inventory feature. Cozi comes closer there with a solid shopping list system, but this side of the app is more of an “add tasks manually” sort of affair. What I wanted was a list already filled in with common household products that I could then modify to my exact needs. The other problem I ran into is that some of the better features required a subscription or premium upgrade.

I soon realized that most of what I wanted already existed, just not in any one app. So I set out to merge my favorite features into a single household management app. It’s been months since, and I use my custom app nearly every day. I’ve gone through many iterations now, but I still consider the whole thing a work in progress. That’s what I love about it. If I find something else I’d like to handle, I don’t have to wait for a developer to decide to add a feature, as I have the power to update my custom app at will.

I was going to make this into a proper Android app from day one using Claude Code. I started by creating a Google Sheet with all the core features I was thinking of and how I wanted it presented; I crafted a reference doc, basically. From there, I dove right in, but it didn’t take long to figure out this was going to be a slow process.

See, Claude can’t generate an Android app out of thin air; it can only help with the coding process. You’ll still need tools like Android Studio, and there’s a lot of manual input work with this method. This didn’t scare me off initially, and I actually got a (sort of) working alpha build out of it.

The problem is that after spending a lot of time building it, it only took me a few minutes of testing to realize it was still missing many features — and that it would take me forever to build at this rate. I soon realized that I needed a way to build rapid prototypes that would let me quickly try out new ideas. So I turned to Claude to make me a web app instead.

Needless to say, this was a much easier, user-friendly method. You don’t need much in the way of extra tools, and a simple back-and-forth vibe coding is all it really takes.

The only downside? The experience was a bit clunky since I had to open it in a browser directly to use it. That’s when I decided to turn it into a progressive web app. This meant using Claude to generate a few other files like manifest.json (tells Android the app name, icons, and colors) and service-worker.js (handles offline caching).

Once I had a functional prototype ready, I then needed a way to host the files so Chrome could recognize it as a PWA, as it won’t recognize local files. There are a few ways to do this, including using a local server app or hosting the app online via something like GitHub. I chose the latter, which is easy to install (you can even ask Claude for help with this).

The first version of my web app was still missing a lot of features, though it was closer than my earliest attempts with a real Android app. With time, each iteration got closer to what I was actually looking for. Some features were harder to implement than others, so that also meant learning new skills, new tools, and just getting better at thinking outside the box here.

Syncing, for example, was a hard nut to crack. A manual file-based import/export feature was easy enough to add as a stopgap, but that’s a really limited way to use a household app across multiple devices. My first solution was to install the app on a tablet we barely used anymore, then get a magnetic case that let me stick it to the fridge as a family hub.

Within a month or so, I had an app that tracked family spending, inventory, and cleaning tasks, with individual tabs for each family member’s specific chores. I even got Google Calendar integration working and started experimenting with Firebase to build a database for smoother cross-device syncing.

The great thing about working with Claude — even on something as simple as a PWA — is how much you pick up about the app-building process along the way. It starts out pretty plug-and-play, but working around the base limitations of this method gradually makes you more familiar with coding and other tools.

You may eventually hit a wall, though. Back-and-forth edits can burn through your usage limits faster than building a real app with something like Claude Code, even if this route is faster and easier overall. It’s also important you pick the right model. In my case, it was Opus (the version itself changed over time). And even if you push PWAs to their limits, you’re going to find that the general design and UI will always look somewhat generic and less refined than a real app.

You can work around that to some degree, but without true coding, it will always feel a bit amateurish. You also won’t be able to officially publish these apps without major modifications, like wrapping the web app in a shell with something like Capacitor or Cordova.

Now that I’ve spent a few months with my PWA, I’m ready for a full rewrite. I’ve started using Claude Code, Android Studio, and a few other tools to build a real app with web-based syncing and much more. It’s not fully ready, but I do have a test build, and to say it’s a million miles ahead of the original is an understatement.

Some might say I would have been better off going straight to an Android app. I disagree, as I wouldn’t have gotten here without some training wheels first. Instead, a few months in, I’m on roughly my 10th build of the PWA version and have a rough alpha of the native Android version. That kind of rapid prototyping would have been much harder had I gone the traditional route from day one. It’s still not the prettiest or most feature-complete app, but it works well enough for now, and it’s become a fun side project.

Building PWAs made me more confident with coding and development tools, which makes this a better learning process, even if a native Android app will always provide a better experience. The takeaway: if you have a custom problem you want to solve but aren’t quite sure how to get started, using Claude to make a PWA is a great starting point. You’ll learn a lot, and you might just find yourself hooked on digging deeper. At least that was my experience.